5.1 Primary and secondary component metadata
Primary and secondary components by outcome class, with the trials that measured each.
Covariate-adjusted effects on each primary, secondary, and tertiary milk component, by trial and visit, for the pooled intervention arms and for each randomized arm vs. control. In the tables, ATE is the effect in SD units and ATE_unscaled in native units, both with 95% CIs; perc_imp is the difference between the intervention and control means as a percentage of the control mean; pval_adj is FDR-adjusted within outcome group, trial, and visit; pval_adj_global is adjusted across all trials and visits of an outcome group; and asterisks mark P or Q below 0.05, 0.01, and 0.001.
Primary and secondary components by outcome class, with the trials that measured each.
The targeted metabolites (Biocrates MxP Quant 500 panel) by compound class, with the trials that measured each.
Every primary, secondary, and tertiary effect as a volcano plot, forest plot, or table, in SD or native units, filtered by trial, visit, and contrast. The app runs R in the browser, so the first load takes a moment. Open it in a new tab.
Selecting points in the volcano plot (box or lasso select) filters the table beneath it, and the menus filter both. Each point is one metabolite, pooled intervention vs. control; FDR is controlled within each trial and visit, so this pooled view has no FDR line.